Technology Assessment

(1)

Technology Assessment

Technology

Assessment Program

Prepared for:

Agency for Healthcare Research and Quality 540 Gaither Road

Rockville, Maryland 20850

Comparative evaluation of

radiation treatments for

clinically localized prostate

cancer: an update

August 13, 2010

(2)

Radiation Therapy for Localized Prostate Cancer: an

Update

Technology Assessment Report

Project ID: CANT1209

August 13, 2010

Tufts Evidence-based Practice Center

Stanley Ip, M.D. Tomas Dvorak, M.D. Winifred W. Yu, M.S., R.D. Kamal Patel, M.P.H., M.B.A. Ndidiamaka Obadan, M.D., M.Sc.

Mei Chung, M.P.H. Raveendhara R. Bannuru, M.D.

Joseph Lau, M.D.

This report is based on research conducted by the Tufts EPC under contract to the Agency for Healthcare Research and Quality (AHRQ), Rockville, MD (Contract No. 290 2007 10055 I). The findings and conclusions in this document are those of the author(s) who are responsible for its contents; the findings and conclusions do not necessarily represent the views of AHRQ. No statement in this article should be construed as an official position of the Agency for Healthcare Research and

Quality or of the U.S. Department of Health and Human Services.

The information in this report is intended to help health care decision-makers; patients and clinicians, health system leaders, and policymakers, make

(3)

well-informed decisions and thereby improve the quality of health care services. This report is not intended to be a substitute for the application of clinical judgment. Decisions concerning the provision of clinical care should consider this report in the same way as any medical reference and in conjunction with all other pertinent information, i.e., in the context of available resources and circumstances presented by individual patients.

This report may be used, in whole or in part, as the basis for development of clinical practice guidelines and other quality enhancement tools, or as a basis for reimbursement and coverage policies. AHRQ or U.S. Department of Health and Human Services endorsement of such derivative products may not be stated or implied.

The investigator, Tomas Dvorak, M.D. discloses his affiliation as a

Communications Committee Member for the American Society for Radiation Oncology. This is a voluntary position. Dr. Dvorak does not receive any financial remuneration from participating on this committee.

All other investigators do not have any affiliation or financial involvement related to the materials presented in this report.

(4)

Peer Reviewers

We wish to acknowledge individuals listed below for their review of this report. This report has been reviewed in draft form by individuals chosen for their expertise and diverse perspectives. The purpose of the review was to provide candid, objective, and critical comments for consideration by the EPC in preparation of the final report. Synthesis of the scientific literature presented here does not necessarily represent the views of

individual reviewers. Patrick A. Kupelian, MD Oncologist

MD Anderson Cancer Center Orlando Orlando, FL

Colleen A. Lawton, MD, FACR, FASTRO Professor of Radiation Oncology

Medical College of Wisconsin Milwaukee, WI

W. Robert Lee, MD, Med, MS Radiation Oncologist

Duke University Medical Center Durham , NC

(5)

Table of Contents for Tables

Table 1. Strength of evidence for radiation treatments of clinically localized prostate cancer...ES-3 Table 2. Comparison of external beam radiation therapy (EBRT) modalities……….……7

Table 3. Strength of evidence for radiation treatments of clinically localized prostate cancer…...16

Table 4. Comparisons with Moderate level of evidence……….….17

Table 5. Genitourinary toxicity: LDRBT vs. EBRT………20

Table 6. Gastrointestinal toxicity: LDRBT vs. EBRT……….…20

Table 7. Genitourinary toxicity: combinations of radiotherapies vs. EBRT or LDRBT……….…22

Table 8. Gastrointestinal toxicity: combinations of radiotherapies vs. EBRT or LDRBT…….….22

(7)

xi

Table 10. Gastrointestinal toxicity: EBRT dose comparisons……….……26 Table 11. Genitourinary toxicity: EBRT fraction size comparisons……….…...27 Table 12. Gastrointestinal toxicity: EBRT fraction size comparisons……….27 Table 13. Effects of patient characteristics on treatment outcomes of different radiation

therapies………...……29

Table of Contents for Figures

Figure 1. Overview of radiation therapy modalities……….………7 Figure 2. Number of comparative primary studies on radiation treatments for clinically localized prostate cancer……….….…...15 Figure 3. Patient survival: radiation therapy vs. no treatment or no initial treatment……….18 Figure 4. Freedom from biochemical failure (5 years of follow-up): LDRBT vs. EBRT….……..19 Figure 5. Freedom from biochemical failure: EBRT dose comparisons (>5 years of follow-up)...24

(8)

(9)

viii

Abbreviations

2D-RT conventional (non-3D planned) RT

3D-CRT conformal RT

ADT androgen deprivation therapy AUA American Urological Association bDFS biochemical (PSA) disease free survival BED biological effective dose

bNED Biochemical no evidence of disease; same as bDFS BT Brachytherapy

CRT-PO conformal RT prostate only

EBRT external beam RT (including 2D-RT, 3D-CRT, IMRT) FFBF freedom from biochemical failure

Gy Gray (unit of radiation dose)

GyE Gray equivalents (used for particle therapy dose reporting) HDR or HDRBT high dose rate brachytherapy

IGRT image guided RT (including IMRT and SBRT) IMRT intensity modulated RT

LDR or LDRBT low dose rate brachytherapy

nADT Neo-adjuvant androgen deprivation therapy NCCN National Comprehensive Cancer Network NCI National Cancer Institute

NT No treatment or no initial treatment

PSA Prostate-specific antigen

PT proton or particle therapy

QoL quality of life

PPI Permanent Prostate Implant Brachytherapy

RT radiation therapy

RTOG Radiation Therapy Oncology Group

SBRT stereotactic body RT

SEER Surveillance, Epidemiology, and End Results Program of the NCI TURP transurethral resection of the prostate

(10)

Executive Summary

Background

The Coverage and Analysis Group at the Centers for Medicare and Medicaid Services (CMS) requested this report from The Technology Assessment Program (TAP) at the Agency for

Healthcare Research and Quality (AHRQ). AHRQ assigned this report to the following Evidence-based Practice Center: Tufts EPC (Contract No. 290 2007 10055 I).

Prostate cancer is the most common noncutaneous malignancy diagnosed in men in the United States. The vast majority of patients diagnosed today have clinically localized prostate cancer (T1-T2N0), which is the subject of this report. A Comparative Effectiveness Review of Therapies for Clinically Localized Prostate Cancer was undertaken on behalf of the Agency for Healthcare Research and Quality (AHRQ) by the Minnesota Evidence-based Practice Center (EPC) in 2007 (Wilt et al. Comparative effectiveness of therapies for clinically localized prostate cancer. Comparative Effectiveness Review No. 13, prepared by Minnesota Evidence-based Practice Center under contract no. 290-02-0009 Rockville, MD: Agency for Healthcare Research and Quality, February 2008. Available at effectivehealthcare.ahrq.gov/reports/final.cfm). The report concluded that “No one therapy can be considered the preferred treatment for localized prostate cancer due to limitations in the body of evidence as well as the likely tradeoffs an individual patient must make between estimated treatment effectiveness, necessity, and adverse effects. All treatment options result in adverse effects (primarily urinary, bowel, and sexual), although the severity and frequency may vary between treatments. Even if differences in

therapeutic effectiveness exist, differences in adverse effects, convenience, and costs are likely to be important factors in individual patient decision making.” As more studies on radiation

treatments have been published since the Minnesota report, the Centers for Medicare and Medicaid Services (CMS) is interested in an update. After consultation with AHRQ and CMS, this technology assessment has been commissioned specifically to examine the recent

comparative studies on radiation treatments of clinically localized prostate cancer.

Methods

This report addressed the following key questions:

1. What are the benefits and harms of radiation therapy for clinically localized prostate cancer compared to no treatment or no initial treatment (watchful waiting, active surveillance, or observation)in terms of clinical outcomes?

2. What are the benefits and harms of different forms of radiation therapy for clinically localized prostate cancer in terms of clinical outcomes? The comparisons of interest are between the following radiation modalities: stereotactic body radiation therapy (SBRT, including CyberKnife® therapy), classically fractionated external beam radiation therapy (EBRT, including 3D-conformal radiation therapy, intensity modulated radiation therapy, and particle therapy), high dose rate brachytherapy (HDRBT), and low dose rate brachytherapy (LDRBT, including permanent brachytherapy).

3. How do specific patient characteristics, e.g., age, race/ethnicity, presence or absence of comorbidities, preferences (e.g., tradeoff of treatment-related adverse effects vs. potential for disease progression) affect the outcomes of these different forms of radiation therapy?

(11)

We relied on findings from the 2008 comparative effectiveness review of therapies for clinically localized prostate cancer conducted by the Minnesota EPC as a springboard for our review. As the Minnesota review conducted its literature search through mid-September 2007, we conducted our literature search from January 2007 to ensure that all relevant and eligible studies are included. The methods for this technology assessment largely follows the methods suggested in the Methods Reference Guide for Effectiveness and Comparative Effectiveness Reviews, Version 1.0 published by AHRQ (available at

effectiveheealthcare.ahrq.gov/repFiles/2007_10DraftMethodsGuide.pdf).

We included randomized controlled trials and non-randomized direct comparative studies of men with clinically localized disease that reported clinical outcomes for T1 or T2 disease. We excluded single cohort studies, adjuvant, salvage, or post-prostatectomy radiation therapy studies, and studies evaluating androgen deprivation therapy. The intervention of interest was radiation treatment used as a first line treatment of prostate cancer. The treatments included various forms of external beam radiation therapy (intensity-modulated radiotherapy, conformal radiation, stereotactic body radiation including CyberKnife®, and proton beam), and

brachytherapy (permanent seed implantation and high dose rate temporary brachytherapy). The treatments reviewed also included combination radiation therapies, such as external beam radiation therapy with brachytherapy boost. The comparators of interest were no treatment or no initial treatment (including watchful waiting and active surveillance) and alternate forms of radiation therapy. Outcomes of interest included overall and prostate cancer-specific survival, metastatic and/or clinical progression free survival, freedom from biochemical (PSA) failure, quality of life, bowel and urinary toxicities, and sexual dysfunction.

From the included studies, we extracted information on patient samples, radiation treatment characteristics (e.g., type of radiation (proton vs. photon), source of radiation (linear accelerator, Cobalt-60, internally planted radioactive seeds), dose, number of fractions, and manufacturer of device), treatment planning algorithm, outcomes (clinical and biochemical), adverse events, and study design. We used a 3-grade (A, B, C) rating system to rate the quality of the individual study. We also used a 3-category rating system (high, moderate, insufficient) to assess the overall strength of evidence for the outcomes reported in each of the comparisons.

Results and Strength of Evidence

We searched for articles on radiation treatments for prostate cancer published between January 2007 and December 2009 in the MEDLINE® and Cochrane Central database and found 51 out of 1,283 articles that met our inclusion criteria. We also added 9 randomized controlled trials (RCTs) relevant to radiation treatments identified in the Minnesota report to our analysis. A total of 62 articles were included in our review. The table below summarized the strength of evidence for the outcomes reported in each of the comparisons.

(12)

Table 1. Strength of evidence for radiation treatments of clinically localized prostate cancer

High Moderate Insufficient

KQ1. Radiation therapy versus no treatment or no initial treatment

Freedom from Biochemical failure Xa

Disease Specific Survival Xb

Genitourinary/Gastrointestinal Toxicity Xc KQ 2. Different forms or doses of radiation

SBRT versus EBRT Xa

SBRT versus HDRBT Xa

SBRT versus LDRBT Xa

EBRT versus HDRBT Xa

EBRT versus LDRBT

Freedom from Biochemical failure Xb

Disease Specific Survival Xc

Genitourinary/Gastrointestinal Toxicity Xb LDRBT versus HDRBT

Freedom from Biochemical failure Xc

Genitourinary/Gastrointestinal Toxicity Xc Combined RT modality comparisons

Genitourinary/Gastrointestinal Toxicity Xd Intra SBRT comparisons

Genitourinary/Gastrointestinal Toxicity Xc Intra EBRT comparisons

Freedom from Biochemical failure Xe

Genitourinary/Gastrointestinal Toxicity Xe Intra BT comparisons

Freedom from Biochemical failure Xf

Genitourinary/Gastrointestinal Toxicity Xb KQ 3. Patient characteristics related to

radiation treatment outcomes

Baseline risk (Stage, PSA, Gleason score) Xb

Gleason score/PSA levels Xc

a: No study available

b: Results inconsistent across studies c: Only one study

d: Predominantly C quality studies

e: ≥2 B quality RCTs that reported similar results (significant difference or no difference) f: Only one RCT and one retrospective study

(13)

Key Question 1. What are the benefits and harms of radiation therapy for clinically localized prostate cancer compared to no treatment or no initial treatment (watchful waiting, active surveillance, or observation)in terms of clinical outcomes?

The strength of evidence for comparing radiation therapy with no treatment or no initial treatment was rated “insufficient” because available data were all provided by retrospective analyses. The Minnesota review did not identify any RCTs that compared external beam radiation therapy with no treatment or no initial treatment; neither did this update review. Data from three retrospective cohorts showed mostly non-significant improvement in disease-specific patient survival in those who received radiation therapy compared to those who had either no treatment or no initial treatment.

Key Question 2. What are the benefits and harms of different forms of radiation therapy for clinically localized prostate cancer in terms of clinical outcomes? The comparisons of interest are between the following radiation modalities: stereotactic body radiation therapy (SBRT,), classically fractionated external beam radiation therapy (EBRT, including 3D-conformal radiation therapy, intensity modulated radiation therapy, and particle therapy), high dose rate brachytherapy (HDRBT), and low dose rate brachytherapy (LDRBT, including permanent brachytherapy).

There were no comparisons between SBRT and any other radiation modality. There were also no comparisons between EBRT and HDRBT.

LDRBT vs. EBRT

Evidence for the comparative efficacy between LDRBT and EBRT on patient survival was rated “insufficient” as there was only one eligible study in this comparison. This retrospective study suggests that there was no difference in disease specific patient survival comparing LDRBT with EBRT.

Evidence for the comparative efficacy between BT and EBRT on biochemical control was rated “insufficient” because the results were inconsistent across the four B-rated studies in this comparison. While two studies found better biochemical control in the LDRBT group compared to the EBRT group, two studies did not find differences between groups.

Evidence for the comparative efficacy between LDRBT and EBRT for genitourinary and gastrointestinal toxicities was rated “insufficient” because the four B-rated studies did not report consistent results. Two studies did and two studies did not show that LDRBT was associated with significantly more genitourinary toxicity than EBRT. For gastrointestinal toxicity, one study showed that LDRBT was associated with less gastrointestinal toxicity compared with EBRT, the other three studies did not find significant difference between LDRBT and EBRT.

Regarding sexual dysfunction, one study showed significantly better outcomes with LDRBT compared with EBRT, another study also reported better outcomes with LDRBT compared with EBRT but the P value of this study was not reported..

Only one study reported cancer incidence comparing LDRBT with EBRT; this study showed a significantly lower incidence of bladder and rectal cancer with LDRBT compared with EBRT. LDRBT vs. HDRBT

Evidence for the comparative efficacy between HDRBT and LDRBT on biochemical outcome was rated “insufficient” as only one retrospective study provided relevant data.

(14)

Comparing HDRBT using Ir-192 (38 Gy or 42 Gy) with LDRBT using Pd-103 (120 Gy), this study did not find a difference in the 5-year freedom from biochemical failure in the two groups.

Combination Therapies: LDRBT plus EBRT in different doses

Evidence on the comparative efficacy of different combinations of radiation was rated “insufficient”, as there were only a few studies in each of the comparisons.

One study did not find a difference in biochemical failure comparing LDRBT plus EBRT in different doses, while another study did not find a difference in biochemical failure comparing LDRBT plus EBRT versus LDRBT. One study did not find a difference in biochemical or clinical failure comparing EBRT with EBRT plus HDRBT.

Limited data suggest greater genitourinary toxicity in EBRT plus LDRBT versus EBRT, and BT plus EBRT versus EBRT.

Our analysis of the data from a study comparing EBRT plus BT versus EBRT showed a significantly increased rate of second primary cancers and late second primary cancers (≥ 5 years) in the EBRT arm.

In addition to comparing different modalities of radiation therapy with each other, we also reviewed comparative evidence within a given radiation modality.

Intra-SBRT comparisons

Evidence for the comparative efficacy on SBRT was rated “insufficient” as only one study qualified for inclusion in this review. This retrospective study found little difference in bladder and rectal toxicities between those who received 35 Gy in 5 fractions and those who received 36.25 Gy also in 5 fractions.

Intra-EBRT Comparisons

For EBRT dose comparison, “moderate” level of evidence from eight studies suggests that higher dose EBRT is associated with increased rates of freedom from biochemical failure at 5 to 10 years compared to lower dose EBRT.

Data from five studies suggest that there is little or no difference in acute and late genitourinary or gastrointestinal toxicities between higher and lower dose EBRT.

For EBRT fractionation comparison, data from three studies suggest that there is no difference between standard fractionation and hypofractionation arms as tested in the studies for freedom from biochemical failure. There is also little or no difference in gastrointestinal toxicity between arms. One B-rated RCT reported a slightly higher acute genitourinary toxicity in the hypofractionation arm compared with the standard fractionation arm, but there was no difference in late genitourinary toxicity.

Intra-LDRBT comparisons

For LDRBT dose and radionuclide comparison studies, “insufficient” level of evidence from one RCT suggests there is little or no difference between I-125 (144 Gy) and Pd-103 (125 Gy) in terms of freedom from biochemical failure at 3 to 6 years. One analysis found that higher

biological effective dose (BED) (>220 Gy) using either I-125 or Pd-103 may improve the 5-year rate of freedom from biochemical failure compared with lower dose (≤220 Gy) in those with higher risk of prostate cancer progression (Gleason score 8 to 10).

(15)

Key Question 3. How do specific patient characteristics, e.g., age, race/ethnicity, presence or absence of comorbidities, preferences (e.g., tradeoff of treatment-related adverse effects vs. potential for disease progression) affect the outcomes of these different forms of radiation therapy?

There were few studies on the potential effects of different patient characteristics on treatment outcomes, apart from patients’ baseline risk. The strength of evidence for evaluating baseline risk as a modifier of outcomes of radiation therapies was rated “insufficient” because there were limited studies for the comparisons reviewed.

Discussion

Because prostate cancer tends to have a long clinical course typically measured in decades, many studies focused on short term adverse events or biochemical control rather than long term clinical efficacy outcomes like metastases and disease-specific mortality. It should be noted that in the studies reviewed, the event rates for grade 3 or greater urinary or bowel toxicity are so low that any statistically significant differences between treatment arms may not translate into

substantive clinical differences.

Many of the findings reported in this review were inconsistent for each of the outcomes of interest. The studies reviewed showed substantial heterogeneity. Even among patients with T1 or T2 prostate cancer, the underlying risk of prostate cancer progression varies widely. An

important weakness in many of these comparative analyses is that patients were given treatments tailored to their individual risk profile (e.g., patients with low risk prostate cancer tend to be given BT versus those with intermediate risk prostate cancer tend to be given EBRT); this makes it difficult, if not impossible, to assess the comparative efficacies between two forms of radiation treatments as the underlying risk of prostate cancer progression in the two groups of patients may be fundamentally different.

The focus of this review is clinically localized prostate cancer (stages T1 and T2). The majority of the patients in these studies had clinically localized disease (stage T1 and

T2); however, approximately one-third of the studies reviewed included up to 20% of patients with stage T3 or higher disease. Excluding them would lead to a drastically reduced number of qualified studies and may inadvertently discard useful data. Similarly, approximately half of the studies had some patients who received androgen deprivation therapies (ADTs), either as a neoadjuvant, concurrent or adjuvant therapy. Many of the studies that included patients with stage T3 or higher disease or ADTs did not report results stratified by patients’ tumor stage or ADT use. Therefore, we are not always able to draw conclusions on the specific treatment effects of the different forms of radiation alone for clinically localized prostate cancer patients (stage T1 and T2), without contamination of results from patients with stage T3 or higher disease, or without contamination of results from patients also treated with ADTs. How these contaminations would affect the “true” treatment effect estimate of radiation alone in only T1-T2 disease is unpredictable.

Conclusion

Definitive benefits of radiation treatments compared to no treatment or no initial treatment for localized prostate cancer could not be determined because available data were insufficient. Data on comparative effectiveness between different forms of radiation treatments

(16)

(BT, EBRT, SBRT) are also inconclusive whether one form of radiation therapy is superior to another form in terms of overall or disease-specific survival. Studies suggest that higher EBRT dose results in increased rates of long-term biochemical control than lower EBRT dose. EBRT administered as a standard fractionation or moderate hypofractionation does not appear to differ with respect to biochemical control and late genitourinary and gastrointestinal toxicities.

Available data suggest that BT might be associated with an increase in genitourinary toxicity compared with EBRT. BT appears to be largely comparable to EBRT in the rates of

gastrointestinal toxicity. However, more and better quality studies are needed to either confirm or refute these suggested findings.

(17)

Introduction

The Coverage and Analysis Group at the Centers for Medicare and Medicaid Services (CMS) requested this report from The Technology Assessment Program (TAP) at the Agency for

Healthcare Research and Quality (AHRQ). AHRQ assigned this report to the following Evidence-based Practice Center: Tufts EPC (Contract No. 290 2007 10055 I).

Prostate cancer is the most common noncutaneous malignancy diagnosed in men in United States. The American Cancer Society estimates that in 2009, approximately 192,000 men were diagnosed with prostate cancer, accounting for 25% of all new cancer cases, and that

approximately 27,000 men died of the disease.1 Median age at diagnosis is 67 years

(seer.cancer.gov/csr/1975_2006/results_single/sect_01_table.11_2pgs.pdf). However, autopsy studies suggest that 30% of men already have undiagnosed prostate cancer by age 40 and as many as 70-80% may have clinically “silent” prostate cancer by age 85.2 It has been estimated that approximately 50% of men undergo a routine prostate specific antigen (PSA) screening. Widespread PSA testing has doubled the incidence of prostate cancer, and results in the lifetime risk of prostate cancer of approximately 16% (seer.cancer.gov/statfacts/html/prost.html).

In addition to increasing incidence, PSA screening is also changing the characteristics of diagnosed prostate cancer. Data from the CaPSURE registry containing 8,685 men with biopsy-proven prostate cancer showed that the incidence of clinically “silent” T1 tumors (tumors

diagnosed incidentally during transurethral resection of prostate for benign prostatic hypertrophy or tumors diagnosed by PSA screening, without clinical evidence by digital rectal exam)

increased from 17% in 1989 to 48% in 2001.3 In the Prostate, Lung, Colorectal, and Ovarian Screening Trial (PLCO), 95% of patients were diagnosed with clinically localized disease, while only 1.6% were diagnosed with locally advanced disease, and 2.4% were diagnosed with

metastatic disease.4 Overall, the vast majority of patients diagnosed today have clinically localized prostate cancer (T1-T2N0), which is the subject of this report.

To appreciate the impact of treatment interventions, it is important to understand the natural history of untreated clinically localized prostate cancer. Our understanding of this process is limited because of the stage shift to earlier disease with PSA screening discussed above. Investigators from a European prostate cancer screening trial (ERSPC) have estimated that the mean lead time bias for screening-detected cancers versus clinically-detected cancers could be 11.2 years, with an estimated overdetection rate of 50%.5 This suggests that after PSA diagnosis of prostate cancer (Stage T1c), it can take more than 10 years before the disease becomes clinically apparent. The natural history of prostate cancer diagnosed clinically is better

characterized based on data from pre-PSA era. A cohort study from Sweden tracked 223 patients with localized prostate cancer diagnosed between 1977 and 1984, with a median follow-up of 21 years.6 Most of the cancers had an indolent course during the first 15 years of follow-up, with progression-free survival of 45%, distant metastasis-free survival of 77%, and prostate cancer-specific survival of 79%. However, between 15 and 20 years of follow-up, there was a

significant decrease in progression-free survival to 36%, distant metastasis-free survival to 51%, and prostate cancer-specific survival of 54%. The authors concluded that most prostate cancer patients diagnosed clinically at an early stage have an indolent course, but aggressive metastatic disease may develop in the long term. A similar cohort study from Connecticut tracked 767 patients with localized prostate cancer diagnosed between 1971 and 1984.7 Their 20-year prostate cancer-specific survival was 71% and overall survival was 7%. The 20-year cancer-specific

(18)

survival for patients with a low grade (Gleason Score 2-6) was 81%, with intermediate grade (Gleason Score 7) was 55%, and with high grade (Gleason Score 8-10) was 34%. In contrast to the Swedish study, there was no worsening of cancer-specific survival after 15 years.

Extrapolating from these studies, the natural history for an average 70 year old patient diagnosed today could result in the development of clinically evident disease in 10 years and a 50% chance of survival from prostate cancer in 30 years after diagnosis, though the rate would be dependent on initial grade of the tumor. To put these numbers into perspective, for that average 70 year old man the probability of survival for 10 years (when he would develop clinically evident disease) is 65% and the probability of survival for another 30 years (when he would have a 50% risk of dying from prostate cancer) is less than 1% (ssa.gov/OACT/STATS/table4c6.html).

Because of the differential survival rates based on tumor grade, there has been an increased focus on identifying and treating patients with aggressive subtypes whose overall survival is likely to be impacted by their cancer, while deferring treatment for patients with indolent subtypes and/or short life-expectancy, whose overall survival is not likely to be impacted by their cancer. Depending on patient’s risk profile, there are numerous treatment options available, which include active surveillance (deferred initial therapy; with continued surveillance and predetermined action levels that will trigger definitive therapy), watchful waiting (either the same as active surveillance or as no definitive therapy regardless of disease progression, until death), surgery, radiation therapy, cryotherapy, high intensity focused

ultrasound, and androgen deprivation therapy. For the purpose of this report, watchful waiting is considered equivalent to active surveillance, and the No Treatment or No Initial Treatment comparator category in Key Question 1 (see below) includes active surveillance, watchful waiting, and observation. Currently, the National Comprehensive Care Network guidelines represent a standard of care in United States, and the NCCN prostate cancer guideline outlines which treatment options may be appropriate for which patients

(nccn.org/professionals/physician_gls/PDF/prostate.pdf).

A Comparative Effectiveness Review of Therapies for Clinically Localized Prostate Cancer was undertaken on behalf of the Agency for Healthcare Research and Quality (AHRQ) by the Minnesota Evidence-based Practice Center (EPC) in 2007 (Wilt et al. Comparative effectiveness of therapies for clinically localized prostate cancer. Comparative Effectiveness Review No. 13, prepared by Minnesota Evidence-based Practice Center under contract no. 290-02-0009

Rockville, MD: Agency for Healthcare Research and Quality, February 2008. Available at

effectivehealthcare.ahrq.gov/reports/final.cfm). The report concluded that “No one therapy can be considered the preferred treatment for localized prostate cancer due to limitations in the body of evidence as well as the likely tradeoffs an individual patient must make between estimated treatment effectiveness, necessity, and adverse effects. All treatment options result in adverse effects (primarily urinary, bowel, and sexual), although the severity and frequency may vary between treatments. Even if differences in therapeutic effectiveness exist, differences in adverse effects, convenience, and costs are likely to be important factors in individual patient decision making.” As more studies on radiation treatments have been published since the Minnesota report, the Centers for Medicare and Medicaid Services (CMS) is interested in an update. After consultation with AHRQ and CMS, this technology assessment has been commissioned

specifically to examine the recent comparative studies on radiation treatments of prostate cancer. Radiation therapy uses high-energy ionizing radiation to damage DNA of tumor cells,

ultimately causing cell death and resulting in tumor eradication. As part of the radiation

(19)

leading to development of side effects. There are three fundamental questions involved with the delivery of radiation to a tumor: 1) What should be the actual target of radiation? 2) How do we deliver radiation to it most effectively and safely? 3) What dose scheme will we use?

Understanding these questions will help to understand the evolution of radiation technology and current efforts at further advances.

From a technology perspective, there are multiple methods of delivering radiation to the prostate, and these are summarized in Figure 1. Radiation can be delivered from outside the body using man-made accelerators, which is known as teletherapy or external beam radiation therapy (EBRT). Or it can be delivered by implanting naturally radioactive elements directly into the tumor, which is known as brachytherapy (BT). The process of planning a treatment and

delivering radiation therapy is different, depending on whether EBRT or BT is used. For EBRT, the first step involves acquiring a planning CT scan of the patient to outline the tumor. The next step is to virtually plan the treatment using computer models to ensure that radiation is delivered safely and effectively. Finally, the patient comes for the treatment itself, which is typically delivered in multiple sessions called fractions. For BT, treatment planning can be done either prior to the surgical implant in a similar fashion as for EBRT or it can be done inside the operating room at the time of the implant of radioactive sources, usually using ultrasound guidance.

Actual target of radiation

The extent of the irradiation area (field size) is an important issue, because the larger the area treated, the more normal tissues are incidentally irradiated, and the more frequent and severe the side effects (conversely, the smaller the field size, the more likely that the entire tumor will not be adequately treated). Field size is driven both by clinical decisions about the extent of disease, as well as by technical factors regarding prostate visualization and prostate motion. Many technological advances in radiation therapy focus on minimizing the treated volume to minimize toxicity. Prostate cancer typically arises in multiple foci within the prostate gland, and thus far no imaging methods have been able to reliably identify involved areas of the prostate. For now, the entire prostate gland typically serves as the target for radiation. For some prostate tumors, there can be subclinical microscopic extension outside of the prostate into the surrounding tissues, seminal vesicles, and regional lymph nodes. There is no definitive evidence available to guide radiation oncologists on when to treat just the prostate gland, when to irradiate some of the surrounding tissues, when to irradiate the seminal vesicles, and when to treat the lymph nodes. First efforts at targeting the prostate used plain x-rays for visualization. Unfortunately, the prostate gland cannot be reliably distinguished from the surrounding soft tissues on plain films. This method of treatment planning is called two-dimensional planning. Development of CT scans for treatment planning in the 1980’s resulted in better visualization of the prostate gland and the lymphatic drainage. The ability to outline the target in three dimensions led to 3D planning, which is the standard today. However, there is a wide variability among individual radiation oncologists in outlining the shape and location of the prostate gland or the lymphatic drainage.8, 9 MRI scans are able to better distinguish the soft tissue densities of the prostate and peri-prostatic tissue, and are becoming a more common tool for radiation oncologists to delineate the radiation target.

Another factor complicating the determination of tumor location is the fact that the planning CT scan is a momentary snapshot in time that shows only where the prostate gland happens to be during the scan. From day to day (inter-fraction motion), and even from minute to minute during

(20)

treatment (intra-fraction motion), the prostate can move significantly.10 The location of the prostate is determined by a number of factors, including the volume of rectal filling and the amount of bladder filling. If we do not know where the prostate may be at any given moment, we need to treat the entire area where it could potentially be, and thus irradiate unnecessarily

significant amount of normal tissues. There are three broad strategies to deal with the motion of the prostate: 1) improved immobilization of the prostate, 2) increased frequency of localizing (imaging) the prostate over time, 3) implanting radiation directly into the prostate, such that the sources move with the prostate.

There are a number of immobilization techniques or devices employed during external beam radiation therapy, such as abdominal compression, endorectal balloon, foot holders, knee

supports, and pelvic immobilizers, which attempt to more reproducibly fix the position of the prostate. In terms of prostate localization for treatment, patients used to be positioned on the treatment table every day using external skin tattoos, and their position would be verified once per week using bony landmarks on x-ray. Advancements in localization have included using daily imaging prior to each treatment session with ultrasound, x-rays combined with implanted fiduciary markers, and on-board CT scans. This approach eliminates daily (inter-fraction)

variability in prostate position resulting from bladder and rectal filling. However, there is still the issue of intra-fraction mobility during the treatment itself. Two approaches are currently used to address this problem: implanted fiducial markers into the prostate that are tracked continuously during the treatment using electromagnetic fields (Calypso System from Calypso Medical Technologies), or implanted fiducial markers that are tracked prior to each treatment beam every few seconds (CyberKnife® from Accuray Inc., Sunnyvale, CA).

Implanting radiation directly into the target is another method of improving dose delivery. There are two forms: 1) permanent implantation of many radioactive seeds into the prostate, which will deliver their radiation dose over the course of weeks to months, and is known as low dose rate brachytherapy (LDRBT), and 2) temporary implantation of catheters into the prostate, through which radioactive seeds are temporarily placed and deliver their radiation dose over the course of minutes, and is known as high dose rate brachytherapy (HDRBT). The downside of this approach is that it is an operative procedure and that radiation can only be delivered to the prostate, and not easily to seminal vesicles or lymph nodes.

Delivering radiation effectively and safely

Planning the radiation treatment involves generating a virtual model of the patient and mathematically estimating the dose that the tumor and the surrounding normal tissues will

receive during treatment. The dose to each individual patient is actually not known; the estimated dose is verified by irradiating a physical model or an x-ray film and measuring the delivered dose. There are several different models of radiation deposition available. Initially, plain x-rays were used for treatment planning, and the amount of dose given was only calculated for few points within the patient. Using this approach, the amount of radiation actually received by the various parts of the prostate and by the surrounding normal organs was essentially unknown. This process is known as conventional radiation or 2D-radiation. When CT scans became incorporated into the treatment planning process, the prostate gland as well as the surrounding organs could be individually identified. The treatment planning software allowed the calculation of dose at any point in the 3-D space, allowing for a much more precise estimate of dose to the target and to surrounding normal organs. In addition, radiation could be targeted more tightly around the prostate to decrease irradiation of surrounding normal tissues. This process is known

(21)

as 3D-conformal radiation (3D-CRT). The next improvement came with the ability to change the intensity of the radiation beam itself, while the radiation was being delivered. This process is known as intensity modulated radiation therapy (IMRT), and allows much more precise control over where the radiation is deposited. As discussed in the prostate localization section,

incorporation of daily pre-treatment imaging allowed further precision in targeting the radiation, and is known as image-guided radiation therapy (IGRT). IMRT and IGRT are typically used together, such that the position of the beam is adjusted prior to every treatment (IGRT), and the radiation intensity of the beam is modulated once the treatment begins (IMRT). Further

incorporation of various body immobilization systems into IMRT with IGRT, together with increased daily dose, and limiting the number of treatments to one or few, is known as stereotactic body radiation therapy (SBRT)11. An alternative means of delivering radiation therapy using external beam to the prostate involves proton therapy, using exactly the same process as the standard photon EBRT, but using proton particles instead. Despite the technical advances in delivery of external beam radiation, it may not possible to deliver sufficiently high dose without incurring unacceptable normal tissue toxicity. In these instances, efforts are under way to combine EBRT with brachytherapy for dose escalation, either as EBRT + HDRBT or EBRT + LDRBT (Table 1).

Dose schema used in radiation delivery

The unit of absorbed radiation dose is Gray (Gy), which corresponds to absorption of one joule of energy by one kilogram of matter. Historically, dose in prostate cancer was delivered in 1.8 – 2.0 Gy per treatment (fraction). Before development of 3D-CRT (see above), maximum tolerable dose was approximately 70 Gy delivered in 35 fractions over 7 weeks, above which unacceptable toxicity resulted. With technology developments described above, dose escalation over 80 Gy has been employed. Current NCCN guidelines recommend 75.6 to 80+ Gy doses, delivered over 7-8 weeks. There is some evidence that prostate cancer may be better treated with larger doses per fraction (> 2 Gy) than is currently standard.12 To account for the fact that a larger dose per fraction results in dramatically more DNA damage and subsequently cell kill, it is important to convert the physical dose to biologically-equivalent dose (BED).13 Using this concept, a regimen of 7 Gy per fraction for 5 treatments (absolute dose 35 Gy) could result in biologically equivalent tumor control to an IMRT regimen of 2 Gy per fraction for 42 fractions (absolute dose 84 Gy); absolute dose comparisons between different regimens may not be meaningful if the dose per fraction is not the same.

Key questions for this report

Understanding the process of delivering radiation therapy described above is crucial to comprehend this technology assessment. This report addressed the following key questions: 1. What are the benefits and harms of radiation therapy for clinically localized prostate cancer compared to no treatment or no initial treatment (watchful waiting, active surveillance, or observation)in terms of clinical outcomes?

2. What are the benefits and harms of different forms of radiation therapy for clinically localized prostate cancer in terms of clinical outcomes? The comparisons of interest are between the following radiation modalities: stereotactic body radiation therapy (SBRT, including CyberKnife® therapy), classically fractionated external beam radiation therapy (EBRT, including 3D-conformal radiation therapy, intensity modulated radiation therapy, and particle therapy), high dose rate brachytherapy (HDRBT), and low dose rate brachytherapy (LDRBT,

(22)

including permanent brachytherapy). These modalities will be specifically compared with each other, i.e., LDRBT VS. EBRT, HDRBT VS. LDRBT, SBRT vs. EBRT, SBRT vs. HDRBT, SBRT vs. LDRBT, EBRT vs. HDRBT, combination therapies, SBRT comparisons, intra-EBRT comparisons, and intra-BT comparisons.

3. How do specific patient characteristics, e.g., age, race/ethnicity, presence or absence of comorbidities, preferences (e.g., tradeoff of treatment-related adverse effects vs. potential for disease progression) affect the outcomes of these different forms of radiation therapy?

(23)

Figure 1: Overview of radiation therapy modalities Radiation Therapy Teletherapy Brachytherapy External Beam RT (EBRT) Sterotactic Body RT (SBRT) *2D RT *3D CRT *IMRT/IGRT *Protons *"Linac" based *CyberKnife(R) *Protons Low-dose Rate Brachytherapy (LDRBT) High-dose Rate Brachytherapy (HDRBT) *I-125 *Pd-103 *Cs-131 *Ir-192

2D RT is two-dimensional radiation therapy; 3D CRT is three dimensional conformal radiation therapy; IMRT is intensity modulated radiation therapy; IGRT is image-guided radiation therapy; I-125, Pd-103, Cs-131, and Ir-192 are radionuclides used in brachytherapy

Table 2: Comparison of external beam radiation therapy (EBRT) modalities CT Treatment Planning Beam Intensity Modulation Frequent Imaging Stereotactic Immobilization 1-5 Treatment Fractions 2D RT 3D CRT X IMRT X X IGRT X X X SBRT X X X X X Proton Therapy X +/- X X +/- 7

(24)

(25)

Methods

The objective of this technology assessment is to assess, using a systematic review approach, the volume and type of evidence available on radiation treatment for localized prostate cancer. The purpose is to provide a basis for establishing how the research field is evolving and identify areas that may require further research.We relied on findings from the 2008 comparative effectiveness review of therapies for clinically localized prostate cancer conducted by the Minnesota EPC (Wilt et al. Comparative effectiveness of therapies for clinically localized prostate cancer. Comparative Effectiveness Review No. 13. (prepared by Minnesota Evidence-based Practice Center under contract no. 290-02-0009) Rockville, MD: Agency for Healthcare Research and Quality, February 2008, available at effectivehealthcare.ahrq.gov/reports/final.cfm) as a springboard for our review. As the Minnesota review conducted its literature search through mid-September 2007, we conducted our literature search from January 2007 to ensure that all relevant and eligible studies are included. The methods for this technology assessment largely follows the methods suggested in the Methods Reference Guide for Effectiveness and

Comparative Effectiveness Reviews, Version 1.0 published by AHRQ (available at

effectiveheealthcare.ahrq.gov/repFiles/2007_10DraftMethodsGuide.pdf). Please note that explanations for abbreviations of frequently used technical terms have been repeated several times throughout the entire document to help clarify highly technical terminologies. See

Abbreviations for a list of abbreviations used for the entire document.

Literature Search Strategy

Our search strategy used the National Library of Medicine’s Medical Subject Headings (MeSH) keyword nomenclature developed for Medline® and adapted for use in other databases. The searches were limited to the English language. The texts of the major search strategies are given in Appendix A.

We searched the Medline and the Cochrane Library from January 2007 to December 2009 for studies involving adults with clinically localized prostate cancer who underwent radiation treatments. We combined search terms or MeSH terms for prostate neoplasm and terms relevant to radiation therapy (e.g., proton beam, particle beam, external beam, radiotherapy, intensity-modulated radiotherapy, brachytherapy). We limited the search to English language studies in adult humans. We included peer reviewed, primary studies of radiation treatment for clinically localized prostate cancer that had reported either clinical or biochemical outcomes. We excluded case reports and conference abstracts. We did not search systematically for unpublished data. Our local domain expert provided additional relevant and eligible citations.

The identified abstracts results were reviewed independently by six reviewers. All abstracts concerning technical aspects of radiation therapy were re-screened by a radiation oncologist.

Study Eligibility Criteria

Our three key questions concern mainly with radiation treatment. We focused only on direct comparative studies for this technology assessment. Because institutions sometime overhaul their radiation treatment in its entirety from one form to another form (e.g., switching from 3D-CRT to IMRT) and therefore direct comparative study between the different forms of treatment within the same institution would not be possible, we have relaxed this criterion to allow such studies as

(26)

long as the patients were consecutively enrolled. We assessed titles and/or abstracts of citations identified from our literature search for inclusion, using the criteria described below. Full-text articles of potentially relevant abstracts were retrieved and a second review for inclusion was conducted by reapplying the inclusion criteria. Results published only as abstracts were not included in our reviews because adequate information is not available to assess the validity of the data and these reports have generally not been peer-reviewed.

Study designs of interest

We included randomized controlled trials and non-randomized direct comparative studies (same institution, contemporaneous or consecutive enrollment is permissible) of men with clinically localized disease and reported clinical outcomes for T1 or T2 disease (no less than 80% T1 or T2 if the study reported a mixed population including T3 and T4).

We did not place sample size and follow-up length restrictions on these comparative studies. We excluded single cohort studies, adjuvant, salvage, or post-prostatectomy radiation therapy studies, and studies evaluating androgen deprivation therapy in conjunction with radiation

therapy.

Population and condition of interest

We included studies of men with clinically localized prostate cancer (T1-T2, N0-X, M0-X) regardless of age, histologic grade, PSA level, or whether they received hormonal treatments (provided the hormonal treatment was not a standard part of the overall treatment plan in one or both of the treatment arms). If the study did not clearly report T staging, we included the study only if it explicitly stated that it enrolled exclusively patients with localized or low risk disease.

Interventions of interest

The intervention of interest was radiation treatment. The radiation treatment had to be used as a first line treatment of prostate cancer. The treatments included external beam radiation therapy (conformal radiation, intensity-modulated radiotherapy, proton therapy) stereotactic body radiation, and brachytherapy (low dose rate permanent seed implantation and high dose rate temporary brachytherapy). We also included combination radiation therapies, such as external beam radiation therapy with brachytherapy boost. For studies that also examined watchful waiting, active surveillance, or observation, because the study investigators did not use consistent definitions for these terms (e.g., some defined watchful waiting as no definitive treatment ever and some defined it as definitive treatment if disease progresses), we accepted all the different definitions and considered all three to be equivalent and grouped all three into “no treatment or no initial treatment” group.

Comparators of interest

The comparators of interest were no treatment or no initial treatment (including watchful waiting and active surveillance) and alternate forms of radiation therapy.

Outcomes of interest

Outcomes of interest included overall survival, prostate cancer-specific survival, biochemical (PSA), metastatic and/or clinical progression free survival, health status, and quality of life. Adverse events included anticipated events such as bowel, bladder, and sexual dysfunction and unanticipated events.

(27)

Data Extraction

Data from each study were extracted by one of the reviewers and confirmed by another. The extracted data included information on patient samples, radiation treatment characteristics (e.g., type of radiation (proton vs. photon), source of radiation (linear accelerator, Cobalt-60, internally planted radioactive seeds), dose, number of fractions, and manufacturer of device), treatment planning algorithm, outcomes (clinical and biochemical), adverse events, and study design. For most outcomes, 6 months, 12 months, and/or only data from the last reported time point were included. We also evaluated potential sources of bias in the included studies with respect to adequate power, randomization, allocation concealment, intention to treat, adequate length of follow-up, number of dropouts and lost to follow-up. To minimize the possibility of between-EPC (Tufts and Minnesota) differences in interpretation of study findings, we elected to extract the data ourselves from those RCTs identified by the Minnesota report that were of relevance to this review.

Quality Assessment

We used predefined criteria to grade study quality as A, B, or C. This system defines a generic grading system that is applicable to varying study designs including RCTs, nonRCTs, and observational studies. For RCTs, we mainly considered the methods used for randomization, blinding, as well as the use of intention-to-treat analysis, the report of dropout rate and the extent to which valid primary outcomes were described and how well they were reported. For nonRCTs and observational studies, the following elements were considered in assessing quality: clear reporting of eligibility criteria, similarity of comparative groups in terms of baseline

characteristics and prognostic factors, reporting on crossovers, differential loss to follow-up between the comparative groups or overall high loss to follow-up, adjustment for potential confounders, and validity and adequacy of the description of outcomes and results.

A (low risk of bias)

Studies rated “A” have the least bias and results are considered valid. These studies adhere mostly to the commonly held concepts of high quality including the following: a formal randomized controlled study; clear description of the population, setting, interventions, and comparison groups; appropriate measurement of outcomes; appropriate statistical and analytic methods and reporting; no reporting errors; less than 20 percent dropout; clear reporting of dropouts; and no obvious bias.

B

Studies rated “B” are susceptible to some bias, but not sufficient to invalidate the results. They do not meet all the criteria in category “A”. The study may be missing information, making it difficult to assess limitations and potential problems.

C (high risk of bias)

Studies rated “C” have significant bias that may invalidate the results. These studies have serious errors in design, analysis, or reporting; there are large amounts of missing information, or major discrepancies in reporting.

(28)

Rating the Body of Evidence

We assigned an overall grade describing the strength of evidence for each key question that was based on the number and quality of individual studies, duration of follow-up and the consistency across studies. The grades corresponded to the following definitions:

High – High confidence that the evidence reflects the true effect. Further research is very

unlikely to change our confidence in the estimate of effect. There is a high level of assurance with validity of the results for the key question based on at least two high quality studies with long-term follow-up of a relevant population. There is no important scientific disagreement across studies in the results for the key question.

Moderate – Moderate confidence that the evidence reflects the true effect. Further

research may change our confidence in the estimates of effect and may change the estimate. There is a moderate level of assurance with validity of the results for the key question based on fewer than two high quality studies or in high quality studies that lack long-term outcomes of relevant populations. There is little disagreement across studies in the results for the key question.

Insufficient− Evidence is either unavailable or if available, a low level of assurance with

validity of results for the key question. There could be disagreement across studies in the results for the key question.

The grades provide a shorthand notation of the strength of evidence supporting the answers to the key questions. However, they may oversimplify the many complex issues involved in appraising a body of evidence. The individual studies involved in formulating the composite grade differed in their design, reporting, and quality. As a result, the strengths and weaknesses of the individual reports addressing each key question would also be considered, as described in detail in the text and tables.

Data Synthesis

For key question 1 (radiation treatment vs. no treatment or no initial treatment) and key question 2 (comparing different forms of radiation treatment), eligible studies were compiled into sets of summary tables that succinctly present the study features including design, patient-level and intervention-patient-level characteristics, results, and study quality. For summarizing adverse events, as most studies used the Radiation Therapy Oncology Group adverse event classification scheme (rtog.org/members/toxicity/ctcmanual.html; rtog.org/members/toxicity/tox.html) in reporting genitourinary and gastrointestinal toxicities after radiation treatments, we have elected to enumerate only grade 3 or greater events in our summary as they are clinically much more serious than grade 1 or grade 2 events. We also used the grade 3 or greater results as proxy for all the toxicity events in the given category. We did not define acute and late adverse events in this review. We accepted definitions that were used in the individual studies (acute events were variably defined as those that occurred less than 3 to less than 6 months after radiation treatments; late events were therefore variably defined as those that occurred more than 3 to more than 6 months after radiation treatments). Result synthesis is presented in the main body of the report. Detailed results of the individual studies are presented in Appendices C to I.

(29)

13

Forest plot

When more than one study provided sufficient data for calculating the effect sizes, we used forest plot to illustrate the relative strength of treatment effects across a set of studies that addressed the same research question (i.e., the same outcome and radiation therapy

comparisons). For clinical outcomes, we employed the risk or rate difference (RD) as the metric of choice to quantify the effect size. For each study in the figure, the forest plot shows the effect size (represented by a square) and confidence intervals (represented by horizontal lines). A confidence interval crossing the vertical line of zero effect size indicates a non-statistically significant result. We noted at the bottom of the forest plot the interpretation of the risk or rate difference (e.g., favors LDRBT or favors HDRBT) on left and right side of the plot.

Studies that reported continuous outcomes (i.e., sexual dysfunction score, urinary dysfunction score, bowel dysfunction score, and quality of life score) are not included in the forest plots. Detailed results for these outcomes are summarized in various tables located in the appendices.

(30)

(31)

Result Synthesis and Strength of Evidence

We searched for articles on radiation treatments for prostate cancer published between January 2007 and December 2009 in MEDLINE® and Cochrane Central database and found 1,283 relevant citations. Abstract screening of these citations identified 165 potentially relevant articles. Full-text screening of these 165 articles identified 53 that met our eligibility criteria. We also added nine RCTs relevant to radiation treatments identified in the Minnesota report to our analysis. A total of 62 articles were included in our review.

The grand overview figure (Figure 2) detailed how many studies reported an outcome (either as a primary or secondary outcome) that is of interest. The total number of studies is greater than the number of unique studies as each study may have provided data for more than one outcome. Table 3 summarized the strength of evidence for the outcomes reported in each of the

comparisons. Table 4 summarized only those comparisons with moderate level of evidence. Figure 2. Number of comparative primary studies on radiation treatments for clinically localized prostate cancera

a _{Because no studies that compared between SBRT and EBRT, SBRT and HDR, SBRT and LDR, or EBRT vs.}

HDRBT were identified, these comparisons are not listed in this figure.

Table 3. Strength of evidence for radiation treatments of clinically localized prostate cancer

High Moderate Insufficient

KQ1. Radiation therapy versus no treatment or no initial treatment

Freedom from Biochemical failure Xa

(32)

a: No study available

Disease Specific Survival Xb

GU/GI Toxicity Xc

KQ 2. Different forms or doses of radiation

SBRT versus EBRT Xa

SBRT versus HDRBT Xa

SBRT versus LDRBT Xa

EBRT versus HDRBT Xa

EBRT versus LDRBT

Freedom from Biochemical failure Xb

GU/GI Toxicity Xb

LDRBT versus HDRBT

GU/GI Toxicity Xc

Combined RT modality comparisons

GU/GI Toxicity Xd

Intra SBRT comparisons

GU/GI Toxicity Xc

Intra EBRT comparisons

Freedom from Biochemical failure Xe

GU/GI Toxicity Xe

Intra BT comparisons

Freedom from Biochemical failure Xf

GU/GI Toxicity Xb

KQ 3. Patient characteristics related to radiation treatment outcomes

Baseline risk (Stage, PSA, Gleason score) Xb

Gleason score/PSA levels Xc

b: Results inconsistent across studies c: Only one study

d: Predominantly C quality studies

e: ≥2 B quality RCTs that reported similar results (significant difference or no difference) f: Only one RCT and one retrospective study

(33)

Table 4. Comparisons with Moderate level of evidence

Biochemical failure Genitourinary toxicity Gastrointestinal toxicity

EBRT dose comparisons Xa Xb Xb

EBRT fraction comparisons Xc _Xc _Xc

a _{Higher dose EBRT is associated with increased rates of freedom from biochemical failure at 5 to 10}

years compared to lower dose EBRT

b _{Little or no difference between higher and lower dose EBRT}

c _{Little or no difference between hypofractionation and standard fractionation arms}

Key question 1: radiation therapy versus no treatment or no initial treatment

(Figure 3)

The strength of evidence for comparing radiation therapy with either no treatment or no initial treatment was rated “insufficient” as all five eligible studies were retrospective analyses.

14-18_{These B-rated studies mostly provided data on different health outcomes. This makes it}

difficult to draw adequate conclusions regarding the same health outcome from the aggregate of the few qualified studies. The Minnesota review did not identify any RCTs that compared external beam radiation therapy with watchful waiting; neither did this update review.

Data from three retrospective cohorts showed mostly non-significant improvement in disease-specific survival in those patients who received radiation therapy compared with those who had either no treatment or no initial treatment.15-17

Only one study reported genitourinary toxicity outcome18 and found no difference between BT or EBRT and no treatment or no initial treatment, but higher rate of receiving urethral stricture treatment in patients treated with combined EBRT and BT, compared with those with no treatment or no initial treatment. One study reported incidence of second primary cancer,14

and found significantly higher rates of second primary cancer in patients treated with EBRT compared with those with no treatment or no initial treatment, but no difference between patients treated with BT and those with no treatment or no initial treatment.

Figure 3. Patient survival: radiation therapy vs. no treatment or no initial treatment

(34)

Albertsen (2007) Tewari (2007) Zhou (2009) Zhou (2009) ID Study 0.67 (0.39, 1.15) 0.64 (0.38, 1.07) 0.45 (0.23, 0.88) 0.66 (0.41, 1.05) adjRR (95% CI) adjHR or EBRT vs. Observation Radiation vs. Watchful waiting BT vs. No treatment* EBRT vs. No treatment* comparison B B B B quality 0.67 (0.39, 1.15) 0.64 (0.38, 1.07) 0.45 (0.23, 0.88) 0.66 (0.41, 1.05) adjRR (95% CI) adjHR or EBRT vs. Observation Radiation vs. Watchful waiting BT vs. No treatment* EBRT vs. No treatment* comparison

Favors EBRT or BT Favors Obs or No Tx 1

.2 .5 .8 1 1.5 2 4

Prostate cancer specific survival rate

Key question 2: different forms of radiation therapies

SBRT vs. other radiation modalities; EBRT vs. HDRBT

There were no comparisons found between SBRT and any other radiation modality. There were also no comparisons between EBRT and HDRBT.

LDRBT versus EBRT (Tables 5, 6; Figure 4)

The strength of evidence for the comparative efficacy between LDRBT and EBRT on disease specific patient survival was rated “insufficient” as there was only one eligible study in this comparison. This B-rated retrospective comparison did not find a difference in disease specific patient survival comparing LDRBT with EBRT.17

The strength of evidence for the comparison between LDRBT and EBRT on biochemical control was rated “insufficient” because the results were inconsistent across the four B-rated retrospective studies.19-22_{While two studies found better biochemical control in the LDRBT}

group compared with the EBRT group,21, 22 two studies did not find differences between groups.19, 20 (N.B. only three B studies were depicted in Figure 4, one B study was not depicted because it did not provide crude rates20).

(35)

Figure 4. Freedom from biochemical failure (5 years of follow-up): LDRBT vs. EBRT† . . [1] no ADT Pe (2009) [2] with ADT Wong (2009) Pickles (2010) Gondi (2007) Jabbar (2010) ID Study -0.01 (-0.05, 0.03) -0.13 (-0.18, -0.09) -0.10 (-0.17, -0.03) -0.17 (-0.29, -0.05) 0.01 (-0.06, 0.07) RD (95% CI) 162/171 473/584 118/139 90/141 115/124 EBRT Events, 181/189 212/225 132/139 58/72 123/134 LDRBT Events, B B B C C quality -0.01 (-0.05, 0.03) -0.13 (-0.18, -0.09) -0.10 (-0.17, -0.03) -0.17 (-0.29, -0.05) 0.01 (-0.06, 0.07) RD (95% CI) 162/171 473/584 118/139 90/141 115/124 EBRT Events,

Favors LDRBT Favors EBRT 0

-.3 -.2 -.1 -.05 -.02 0 .02 .05 .1 .2

Outcome: freedom from biochemical failure (5 years of followup)

*Gondi (2007) comparing BT (with or without ADT) with EBRT (without ADT).

† Eade (2008) was not depicted here because only the actuarial rates and not the crude rates were reported.

Evidence for the comparative efficacy between LDRBT and EBRT for genitourinary and gastrointestinal toxicities was rated “insufficient” because the two prospective23, 24 and two retrospective20, 22 B-rated studies did not report consistent results. Two studies showed that LDRBT was associated with significantly more late genitourinary toxicity than EBRT, but no difference in acute toxicity20, 22 one study showed that LDRBT was associated with more genitourinary toxicity compared with EBRT but the result was non-significant23 and one study did not find significant difference in genitourinary toxicity between LDRBT and EBRT.24 For gastrointestinal toxicity, one study showed that LDRBT was associated with less gastrointestinal toxicity compared with EBRT;24_{the other three studies did not find significant difference}

between LDRBT and EBRT.20, 22, 23

Regarding sexual dysfunction, one B-rated prospective cohort study showed significantly better outcomes with LDRBT compared with EBRT,24 another B-rated prospective cohort study also reported better outcomes with LDRBT compared with EBRT, but the P value of this result was not reported.23

Only one study reported cancer incidence.25 This B-rated retrospective cohort study showed a significantly lower incidence of bladder and rectal cancer with LDRBT compared with EBRT.